I am applying for a National Institute of Mental Health (NIMH) K23 Mentored Patient-Oriented Research Career Development Award to receive the training and research experience needed to become an independent investigator. My long-term goal is a career in academic psychiatry focused on research that improves the care of patients with bipolar disorder. My plan is to conduct epidemiologic studies that clarify the relative safety and effectiveness of pharmacotherapy for this serious illness. The ability to conduct large-scale epidemiologic studies of comparative drug safety and effectiveness will complement my current strengths- solid training and professional experience in the care of patients with bipolar disorder across a variety of settings and a strong foundation in the design and conduct of clinical trials. Importantly, this ability will also allow me to examine the effects of medicines-both as single treatments and in combinations-across a wider variety of clinically important endpoints than would be possible with clinical trials alone, and in important patient subgroups that are under-represented or excluded from clinical trials. My mentors and I have developed a career development and research program designed to meet this goal. The career development program includes further training in the core disciplines of pharmacoepidemiology, advanced training in the pathophysiology, diagnosis, and management of adult bipolar disorders, and development of general academic research career skills. It also includes training in the use of four important pharmacoepidemiology databases: the Tennessee Medicaid research database (training via my research studies), the FDA's adverse event reporting system (AERS, training via 8 week FDA visit), and the Kaiser-California and HMO Research Network databases (training via 1 year interaction with FDA Medication Exposure in Pregnancy Research Program, or MEPREP). Database training will be enhanced by planned interactions with prominent experts familiar with these databases and visits to the sites where these databases are maintained. My two proposed research studies address important unanswered questions regarding relative safety of pharmacotherapy for bipolar disorder. The first study considers the frequent use of the combination of a mood-stabilizer and atypical antipsychotic for initial maintenance therapy. I will test the hypothesis that the risk of new onset diabetes is greater for combination therapy than for monotherapy in a cohort of 25,000 Tennessee Medicaid patients with a new episode of bipolar treatment. The second study focuses on the fetal safety of bipolar medications, an important question given the large number of women of childbearing age with bipolar disorders. I will compare the risk of both congenital malformations and adverse neonatal outcomes in a cohort of 2,000 Tennessee Medicaid women with mood stabilizer only exposure during pregnancy versus 1,000 with atypical antipsychotic only. Each research study is tailored to provide the basis for a subsequent R01 application. The research and career development plans will enable my transition to an independent academic researcher in psychiatry. PUBLIC HEALTH RELEVANCE: Treatment for bipolar disorder is characterized by an increasingly complex array of medication regimens with very limited objective data to guide clinicians'choices. This project focuses on use of data from a large automated health outcomes database to address the urgent need for better data on: 1) the risk of new onset diabetes related to bipolar disorder maintenance treatment with mood stabilizer + atypical antipsychotic drug combinations vs. monotherapy with either, and 2) the comparative risk of major congenital malformations and adverse neonatal outcomes in infants born to mothers with bipolar disorder who are treated during pregnancy with mood stabilizers vs. antipsychotic drugs. These studies will provide essential data for community practice that cannot be easily or feasibly addressed through randomized controlled trials or conventional post-marketing surveillance.